Alibaba's Qwen team has released Qwen-Audio-3.0-TTS, a new text-to-speech model available in real-time (Flash) and high-quality (Plus) versions. This model introduces advanced features such as fine-grained inline tags for controlling emotion and pacing, natural language control for stylistic generation, support for 16 languages, and the ability to produce clean output from noisy audio. Qwen-Audio-3.0-TTS can also generate long-form audio up to three minutes in a single pass and has achieved the top position on the Artificial Analysis TTS Leaderboard. AI
IMPACT Enhances creative applications like game narration and interactive storytelling with advanced control over speech emotion and pacing.
RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
